CCU: Lesson - Data and AI

A group of workers discuss data.Data and AI

Think About It...

As we begin our next lesson, ask yourself a question:

  • What do you think data has to do with AI?

In this lesson, we will focus on data in other terms commonly used in AI: structured, unstructured, and semi-structured. We will also explore the importance of data, ways data is collected, how data is computed, and how it may be used.

Common Terms Connecting Data with AI

Even if you haven't been exposed to the "fancy" terms we mentioned previously concerning data (qualitative, quantitative, categorical, numerical), you've likely participated in both the collection and analysis of such data. Have you ever written a report in social studies about history or government? Then you've collected and analyzed qualitative and categorical data! Have you had to create a chart using measurements you made in either science or math class? Then you've collected and analyzed quantitative and numerical data!

As we move our discussion from human-collected data to the world of AI, it's important to introduce some new terminology. Some of the most impressive features includes how AI utilizes various types of data, how quickly it computes (computational/computing power), and what actions AI chooses as a result of the data. Common types of data used in AI are:

  • Structured Data: highly organized information (like information presented in a table or list: numbers, dates, categories, etc.)
  • Unstructured Data: lacks specific formatting or structure. Examples include text, images, videos, and audio files (…unless stated examples were captured utilizing a smart device)
  • Semi-structured Data: a mix of both structured and unstructured. It has some organization, but doesn’t fit perfectly into tables. An example is a picture taken with a smart device. While the picture alone is considered unstructured data, the information collected by the smart device, such as time, date, location, and face recognition provides structured data.

Video Lesson

It is important to note that data collection must comply with ethical and legal considerations, including user consent, privacy, and security. Ensuring data quality (accuracy, completeness, etc.) is critical for effective AI training and decision making (Big Idea 3).

Take a moment to watch the video, ‘AI Training Data - Bias’ to learn more about training data and biases.

Video Credit: Code.org

Practice Activity

[CC BY-NC-SA 4.0 Links to an external site.] UNLESS OTHERWISE NOTED | IMAGES: LICENSED AND USED ACCORDING TO TERMS OF SUBSCRIPTION - INTENDED ONLY FOR USE WITHIN LESSON.
fizkes/Shutterstock.com. Image used under license from Shutterstock.com and may not be repurposed.